INDEX

Explanations

function arguments and shift

This neuron detects occurrences of function or script arguments (e.g., the “arguments” object, ARGV, var args, and related parameter‐handling code).

New Auto-Interp

Configuration

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

Negative Logits

 ingat

-1.20

 tirantes

-1.17

iç

-1.14

 promessa

-1.13

 rendus

-1.10

 火鍋

-1.09

 prefeitura

-1.09

 πρέπει

-1.08

 volantes

-1.07

 vind

-1.06

POSITIVE LOGITS

1.35

 obstinate

1.28

1.16

“

1.11

但是在

1.07

龟

1.04

炸

1.03

但在

1.02

Don

0.98

 welchen

0.97

Activations Density 0.003%